Exercise 1:

daily_report <- read_csv(url("https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_daily_reports/09-26-2020.csv")) %>% 
  rename(Long = "Long_") 

dr <- daily_report %>% 
  dplyr::group_by(Country_Region) %>% 
  filter(!is.na(Lat)) %>% 
  summarize(Cases = sum(Confirmed),
            Deaths = sum(Deaths),
            Lat = median(Lat),
            Long = median(Long))

#Use plotly to troubleshoot median solution
#ggplotly(  
#ggplot(dr, aes(x = Long, y = Lat, size = Deaths, text= Country_Region)) +
#  borders("world", colour = NA, fill = "grey90") +
#  theme_bw() +
#  geom_point(shape = 21, color='darkmagenta', fill='darkorchid', alpha = 0.5) +
#  labs(title = 'World COVID-19 Confirmed cases',x = '', y = '',
#       size="Cases (x10,000)") +
#  grey_theme +
#  theme(legend.position = "right") +
#  coord_fixed(ratio=1.5)
#) # Close plotly

#Correct bad median lat/lons
dr$Lat[dr$Country_Region=="US"] <- 37.0902
dr$Long[dr$Country_Region=="US"] <- -95.7129
dr$Lat[dr$Country_Region=="United Kingdom"] <- 55.3781
dr$Long[dr$Country_Region=="United Kingdom"] <- -3.4360
dr$Lat[dr$Country_Region=="Canada"] <- 56.1304
dr$Long[dr$Country_Region=="Canada"] <- -106.3468
dr$Lat[dr$Country_Region=="Russia"] <- 61.5240
dr$Long[dr$Country_Region=="Russia"] <- 105.3188
dr$Lat[dr$Country_Region=="Denmark"] <- 56.2639
dr$Long[dr$Country_Region=="Denmark"] <- -9.5018
dr$Lat[dr$Country_Region=="Australia"] <- -25.2744
dr$Long[dr$Country_Region=="Australia"] <- 133.7751

#Final plot
ggplot(dr, aes(x = Long, y = Lat, size = Deaths/1000, text= Country_Region)) +
  borders("world", colour = NA, fill = "grey90") +
  theme_bw() +
  geom_point(shape = 21, color='darkmagenta', fill='darkorchid', alpha = 0.5) +
  labs(title = 'World COVID-19 Deaths',x = '', y = '',
       size="Deaths (x1,000)") +
  grey_theme +
  theme(legend.position = "right") +
  coord_fixed(ratio=1.5)

Exercise 2:

summary(daily_report$Confirmed) #Get summary stats for new case totals

mybreaks <- c(1, 1000, 10000, 100000, 200000) #Set new breaks


ggplot(daily_report, aes(x = Long, y = Lat, size = Confirmed)) +
  borders("state", colour = "grey50", fill = "grey90") + #made state borders darker
  geom_point(aes(x=Long, y=Lat, size=Confirmed, color=Confirmed),stroke=F, alpha=0.7) +
  scale_size_continuous(name="Cases", range=c(1,20), #removed log trans and increased point range
                       breaks=mybreaks, labels = c("1-999",
                                                   "1,000-9,999", "10,000-99,999",
                                                   "100,000-199,999", "200,000+")) +
  scale_color_viridis_c(option="viridis", name="Cases",
                      breaks=mybreaks, trans="log", #kept log trans for color scale as it did a better job visually
                      labels = c("1-999", "1,000-9,999", "10,000-99,999",
                                 "100,000-199,999", "200,000+"))  +
   #Cleaning up the graph
  
  theme_void() + 
  guides( color = guide_legend()) +
  labs(title = "Modified Anisa Dhana's layout for COVID-19 Confirmed Cases in the US") +
  theme(
    legend.position = "bottom",
    text = element_text(color = "#22211d"),
    plot.background = element_rect(fill = "#ffffff", color = NA), 
    panel.background = element_rect(fill = "#ffffff", color = NA), 
    legend.background = element_rect(fill = "#ffffff", color = NA)
  ) +
  coord_fixed(ratio=1.5)

Exercise 3: I updated the cases by county color scale

ggplot(data = us, mapping = aes(x = long, y = lat, group = group)) + 
  coord_fixed(1.3) + 
  # Add data layer
  borders("state", colour = "black") +
  geom_polygon(data = state_join, aes(fill = Confirmed)) +
  scale_fill_viridis(option = "B", trans = "log10", na.value = "lightgrey",
                       name = "Confirmed \ncases") +
  labs(title = "Number of Confirmed Cases by US County",
       x = "Longitude",
       y = "Latitude") +
  theme_bw()

Exercise 4: I changed the state to Connecticut (home state) and the color scale to the viridis::plasma

daily_report <-   read_csv(url("https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_daily_reports/09-26-2020.csv")) %>% 
  rename(Long = "Long_") %>% 
  filter(Province_State == "Connecticut") %>% 
  group_by(Admin2) %>% 
  summarize(Confirmed = sum(Confirmed)) %>% 
  mutate(Admin2 = tolower(Admin2)) %>% 
  filter(!Admin2=="unassigned") %>%  #remove unassigned cases
  rename(Cases = Confirmed)  #renaming confirmed cases column for better plotly

us <- map_data("state")
ct_us <- subset(us, region == "connecticut")
counties <- map_data("county")
ct_county <- subset(counties, region == "connecticut")
state_join <- left_join(ct_county, daily_report, by = c("subregion" = "Admin2")) 
library(plotly)
ggplotly(
  ggplot(data = ct_county, mapping = aes(x = long, y = lat, group = group)) + 
    coord_fixed(1.3) + 
    # Add data layer
    geom_polygon(data = state_join, aes(fill = Cases), color = "grey40") +
    scale_fill_viridis(option = "plasma",
                         name = "Cases") +
    ggtitle("COVID-19 Cases in CT") +
    # Cleaning up the graph
    labs(x=NULL, y=NULL) +
    theme(panel.border = element_blank()) +
    theme(panel.background = element_blank()) +
    theme(axis.ticks = element_blank()) +
    theme(axis.text = element_blank())
)

Exercise 5